Abstract
Objectives
To characterize social factors associated with sudden unexpected infant death (SUID), we examined distribution of social factors across infant race, ethnicity, cause of death (medical, injury-related, unknown) and investigator-derived death grouping utilizing a multi-jurisdictional case registry.
Methods
Retrospective analysis of Centers for Disease Control and Prevention SUID and Sudden Death in the Young Case Registry for infant deaths occurring 2015–2020, including child death review team-compiled clinical and investigative information
Results
Of 2212 infants who experienced SUID, social factors including caregiver criminal history, substance use, and intimate partner violence were associated with lower odds of medical cause of death versus other causes. Over one-third (n=226, 36%) of infants with injury-related deaths had caregivers with substance use concerns; caregiver substance use was common among SUID overall (n=753, 34%). Utilizing an investigator-derived schema, both injury-related deaths and undetermined deaths were grouped together and social factors compared across that cohort versus deaths from medical causes; no statistically significant differences in social factors were identified across groups.
Conclusion
Utilizing a multi-jurisdictional registry, relationships between social factors and cause of death among infants experiencing SUID were identified; social factors (caregiver criminal history, substance use, intimate partner violence) were associated with lower odds of medical cause of death versus other causes. Distribution of social factors among infants experiencing SUID may contribute to emerging disparities and requires further study; efforts are needed to improve social factor data quality. SUID prevention efforts should address impact of social factors on fatality vulnerability and advocacy for social policy change.
Keywords: sudden unexpected infant death, sudden infant death syndrome, social determinants of health, family vulnerability, social context, fatality prevention, child death review
Introduction
Annually in the U.S., approximately 3700 infants die suddenly, unexpectedly.1 The U.S. Centers for Disease Control and Prevention suggest “sudden unexpected infant deaths” (SUID) as broad, umbrella terminology to refer to all sudden infant deaths, including unintentional injury (including sleep-related suffocation), intentional injury (child maltreatment), sudden infant death syndrome (SIDS), and deaths from other known and unknown causes.1 Sudden deaths ultimately attributed to medical causes following comprehensive death investigation and autopsy, like infectious etiologies, are rarer.1,2 Prevention efforts to reduce SUID have traditionally targeted the sleep environment, discouraging dyad surface sharing and hazardous infant placement on soft bedding, couches, and pillows. While contributing to overall SUID rate decline, recent data suggest disparities across marginalized populations; infants from non-Hispanic black (NHB) families experience SUID 2.3 times more frequently than the general population and 2.8 times more than non-Hispanic white (NHW) peers.2,3
Disproportionate SUID among populations historically experiencing concentrated, adverse social factors raises concern about contribution of social factors to fatality risk and potential need for broadened prevention efforts advocating for social policy change.2–5 However, social factors experienced by families whose infants later die suddenly, unexpectedly remain incompletely understood; relationship of social factors to cause of death is largely unknown.5–7 Prior studies examining social factors have been constrained by limited sample size, use of small regional datasets, or narrow focus on select social conditions, limiting interpretation of results and application to SUID prevention strategy; a substantial gap thus exists in present literature.4,8–12 Therefore, the objectives of this study were to characterize social factors associated with SUID utilizing a multi-jurisdictional case registry, examining the distribution of social factors across infant race and ethnicity, association of social factors with medical cause of death versus injury-related deaths and deaths from unknown causes, and association of social factors with death grouping based on an investigator-derived schema.
Methods
Data Source and Study Population
Data from Centers for Disease Control and Prevention’s SUID and Sudden Death in the Young Case Registry (hereafter referred to as Case Registry), a collaborative population-based, multi-jurisdictional surveillance system with National Institutes of Health, Michigan Public Health Institute, and National Center for Fatality Review and Prevention, was obtained regarding 2235 infants (age 12 months or younger) who died suddenly and unexpectedly between 2015 and 2020 in participating jurisdictions.13 The Case Registry, which includes SUID, sudden cardiac and epilepsy-related deaths, suffocation, and other causes was developed to expand age range, scope, and geographic representation of deceased infants/children reviewed by national, multidisciplinary child death review (CDR) teams and supplement existing fatality review protocols.14,15 For Case Registry inclusion, “sudden” is death within 24 hours of first symptom onset or in-hospital death after out-of-hospital cardiac arrest with resuscitation; “unexpected” is death involving someone in apparent good health with stable chronic condition or acute illness unexpected to cause death.13 Cases involving sudden death are reviewed by CDR teams who compile information from multiple medical and investigative sources; variability in data quality and proportion of missing data for each case are impacted by availability and consistency of information accessible to CDR teams across multiple data sources (parent or caregiver witnesses, law enforcement, etc). Cases are then further adjudicated by an advanced review team comprised of jurisdictional cardiologists, neurologists, pathologists, and subject matter experts who categorize the sudden death according to Case Registry algorithms; case data are then additionally entered into the National Fatality Review Case Reporting System (NFR-CRS).13,16–19
Variables obtained from the NFR-CRS included infant age at death, sex, race and ethnicity (from the infant’s death certificate),20 prematurity, and birth weight. Prematurity was categorized as preterm (≤36 weeks) or term and birth weight as normal (>2500 grams), low (≤2500 grams), or very low (≤1500 grams). Maternal smoking behavior, prenatal care receipt, and medical complications during pregnancy were reviewed; complications included cardiovascular (hypertension, pre-eclampsia), neurologic/psychiatric (seizure, addiction), hematologic, respiratory (asthma), endocrine, gynecologic (uterine bleeding, placental issues), and other conditions. Primary and secondary caregivers (those persons primarily responsible for infant care, custody, and control at time of death)20 were categorized as parent (childbearing, non-childbearing, foster, or adoptive parent), non-parent (friends, institutional staff, other relatives) or unknown/missing. Primary/secondary caregiver ages were dichotomized. Manner of death, whether death occurred in the sleeping environment, supervisor (person responsible for infant care at time of death) sleeping or impairment (by substance use, illness, disability, distraction, or absenteeism), and Case Registry categorization were also reviewed.16,19,20 Causes of death from the NFR-CRS were assigned by CDR teams utilizing the infant’s death certificate. Causes of death included external injury (e.g. asphyxia, drowning, intoxication, fire/burn, bodily force or weapon, other causes), medical cause (asthma/respiratory, infection, cardiovascular, neurologic/seizure disorder, malnutrition/dehydration, other conditions), and undetermined cause (unknown if injury or medical cause).
Social factors from the NFR-CRS among primary/secondary caregivers were obtained, including health care access (insurance, prenatal care receipt barriers), child welfare/criminal justice system involvement (child protective services case at death, out-of-home placement, maltreatment perpetration/victimization), economic stability (employment), substance use (including prenatal exposure history), violence exposure (IPV), and housing (homelessness, incident location). Infant race and ethnicity were considered a social factor as this social construct has contributed to systemic racism and historic health inequities. Selection of social factors was determined by study authors based on literature review, clinical experience, and NFR-CRS data availability. Due to variable proportions of missing data specifically across social factors compiled by CDR teams and entered into NFR-CRS, affirmative responses for a given social factor for either caregiver (eg, at least 1 “yes” for the social factor in the NFR-CRS tool) were considered a positive exposure for statistical analysis; negative responses for both caregivers were considered negative exposure to the social factor. Cases in which both caregiver responses were unknown or 1 was no and other unknown were grouped together; this strategy reduced any possibility that social factors would be over-reported among cases.
Several CDR-determined responses involving NFR-CRS variables were re-categorized for analysis and free-text responses recoded as appropriate into existing variables. For example, infant race and ethnicity were combined and reported as single race and Hispanic origin groups NHB, NHW, Hispanic, other non-Hispanic for analysis purposes. Infants were excluded if missing data related to race, ethnicity, or if Case Registry categorization was unknown/missing, unexplained/incomplete, or excluded from categorization; final sample size was 2212 infants. NFR-CRS variables reported as unknown/missing reflect either blank responses in the NFR-CRS tool or those where a CDR team specified the exposure of interest was unknown. Percent unknown/missing per variable was reported due to varied proportions of known data across the Case Registry-derived dataset.
Statistical Analysis
Data were analyzed using statistical software R, version 4.1.1.21 Cell counts <6 were suppressed to prevent inadvertent decedent identification unless reported as unknown/missing; several variables were combined to meet suppression requirements. Continuous variables were summarized using mean (standard deviation SD) or median (IQR); categorical variables were summarized using frequencies and percentages. Multivariable logistic regression analyses were performed to determine association of social factors with medical cause of death (versus other causes), adjusting for prematurity, birth weight (both risk factors for SUID but also contributory to medical causes of death). Odds ratios (OR) were reported. Chi square test examined social factor distribution across infant race, ethnicity, and cause of death. Model assumptions were checked before data analysis and appropriate measures taken if needed. All tests were 2-tailed at level of significance .05.
To further examine distribution of social factors across infants who experienced SUID, the study sample was sub-divided into 2 groups. Group 1 consisted of deaths categorized as unexplained or explained suffocation in the Case Registry. Group 2 consisted of those explained cardiac, explained neurological, explained other, unexplained possible cardiac, unexplained possible cardiac and sudden unexpected death in epilepsy, and unexplained sudden unexpected death in epilepsy in the Case Registry. In this exploratory analysis, it was hypothesized Group 2 deaths (wherein a medical cause was suspected, leading to its Case Registry categorization) would statistically differ in social factor distribution versus Group 1. While not intended to apportion blame, it was hypothesized Group 1 injury-related or unexplained death risk may be environmentally or socially mediated. Group 1 deaths were hypothesized to have affirmative social factor exposures. Chi square test was used for group comparison; logistic regression analysis determined variables associated with Group 1 grouping.
The Institutional Review Board of our tertiary care pediatric hospital approved this study.
Results
Population Characteristics
Of 2212 infants who experienced SUID, median age at death was 2 months (Table 1). Over half were male (57%, n=1262); most were NHB (n=954, 43%) or NHW (n=878, 40%). One-fourth overall were premature (n=553, 25%); 1 in 5 had low/very low birth weight (n=478, 21%). Most childbearing parents had prenatal care (n=1978, 89%); nearly half overall experienced pregnancy complications (n=974, 44%). Nearly all primary (n=2147, 97%) and secondary (n=1497, 68%) caregivers were parents; most were ≥ age 20 years. For 546 (25%) of infants, secondary caregiver information was unknown/missing.
Table 1.
Population Characteristics and distribution of social factors across infant race and ethnicity 2015–2020a
| Infant race and ethnicity | ||||||
|---|---|---|---|---|---|---|
|
|
||||||
| Hispanic | Non-Hispanic Black | Non-Hispanic White | Other non-Hispanic | Overall | P Value | |
|
|
||||||
| Variables | (n = 241) | (n = 954) | (n = 878) | (n = 139) | (n = 2212) | |
|
| ||||||
| Infant age, months | ||||||
| Mean (SD) | 2.95 (2.42) | 3.04 (2.41) | 2.90 (2.46) | 3.17 (2.62) | 2.98 (2.44) | .504 |
| Median [q1, q3] | 3.00 [1.00, 4.00] | 2.50 [1.00, 4.00] | 2.00 [1.00, 4.00] | 2.00 [1.00, 4.50] | 2.00 (1.00, 4.00] | .606 |
| Infant sex, n (%) | ||||||
| Male | 142 (59) | 532 (56) | 498 (57) | 90 (65) | 1262 (57) | .356 |
| Female | 99 (41) | 422 (44) | 380 (43) | 49 (35) | 950 (43) | |
| Prematurity ≤36 weeks, n (%) | ||||||
| Yes | 58 (24) | 291 (31) | 172 (20) | 32 (23) | 553 (25) | <.001 |
| No | 169 (70) | 642 (67) | 673 (77) | 102 (73) | 1586 (72) | |
| Unknown/missing | 14 (6) | 21 (2) | 33 (4) | 5 (4) | 73 (3) | |
| Infant birth weight, n (%) | ||||||
| Normal, >2500 gm | 164 (68) | 632 (66) | 668 (76) | 86 (62) | 1550 (70) | <.001 |
| Low, ≤2500 gm or very low, ≤1500 gm | 43 (18) | 268 (28) | 138 (16) | 29 (21) | 478 (21) | |
| Unknown/missing | 34 (14) | 54 (6) | 72 (8) | 24 (17) | 184 (8) | |
| Maternal smoking during pregnancy, n (%) | ||||||
| Yes | 56 (23) | 283 (30) | 412 (47) | 45 (32) | 796 (36) | <.001 |
| No | 163 (68) | 616 (65) | 409 (47) | 80 (58) | 1268 (57) | |
| Unknown/missing | 22 (9) | 55 (6) | 57 (6) | 14 (10) | 148 (7) | |
| Prenatal care during pregnancy, n (%) | ||||||
| Yes | 214 (89) | 842 (88) | 803 (91) | 119 (86) | 1978 (89) | .002 |
| No | 12 (5) | 75 (8) | 28 (3) | 12 (9) | 127 (6) | |
| Unknown/missing | 15 (6) | 37 (4) | 47 (5) | 8 (6) | 107 (5) | |
| Medical complications during pregnancy, n (%) | ||||||
| Yes | 92 (38) | 472 (49) | 358 (41) | 52 (37) | 974 (44) | <.001 |
| No | 130 (54) | 399 (42) | 433 (49) | 80 (58) | 1042 (47) | |
| Unknown/missing | 19 (8) | 83 (9) | 87 (10) | 7 (5) | 196 (9) | |
| Primary caregiver, n (%) | ||||||
| Parent | 237 (98) | 922 (97) | 852 (97) | 136 (98) | 2147 (97) | .81 |
| Non-parent | * | * | * | * | 55 (2) | |
| Unknown/missing | * | * | * | * | 10 (1) | |
| Primary caregiver age, n (%) | ||||||
| Young ≤20 years | 44 (18) | 123 (13) | 111 (13) | 8 (6) | 286 (13) | .010 |
| Not young age | 177 (73) | 785 (82) | 709 (81) | 118 (85) | 1789 (81) | |
| Unknown/missing | 20 (8) | 46 (5) | 58 (7) | 13 (9) | 137 (6) | |
| Secondary caregiver, n (%) | ||||||
| Parent | 186 (77) | 542 (57) | 659 (75) | 110 (79) | 1497 (68) | <.001 |
| Non-parent | 15 (6) | 96 (10) | 48 (5) | 10 (7) | 169 (8) | |
| Unknown/missing | 40 (17) | 316 (33) | 171 (19) | 19 (14) | 546 (25) | |
| Secondary caregiver age, n (%) | ||||||
| Young ≤20 years | 18 (8) | 44 (5) | 45 (5) | 6 (4) | 113 (5) | <.001 |
| Not young age | 150 (62) | 497 (52) | 574 (65) | 95 (68) | 1316 (59) | |
| Unknown/missing | 73 (30) | 413 (43) | 259 (29) | 38 (27) | 783 (35) | |
| SDY death categorization, n (%) | ||||||
| Explained cardiac | * | * | * | * | 31 (1) | .962 |
| Explained neurologic | * | * | * | * | * | |
| Explained suffocation | 67 (28) | 257 (27) | 271 (31) | 38 (27) | 633 (29) | |
| Explained other | 20 (8) | 87 (9) | 64 (7) | 12 (9) | 183 (8) | |
| Unexplained possible cardiac | * | * | * | * | * | |
| Unexplained possible cardiac and SUDEP | * | * | * | * | * | |
| Unexplained SUDEP | * | * | * | * | * | |
| Unexplained death | 138 (57) | 561 (59) | 491 (56) | 78 (56) | 1268 (57) | |
| Manner of death, n (%) | ||||||
| Natural | 34 (14) | 135 (14) | 112 (13) | 27 (19) | 308 (14) | .011 |
| Accident or homicide | * | * | * | * | * | |
| Unknown/undetermined/pending | 140 (58) | 574 (60) | 491 (56) | 66 (47) | 1271 (58) | |
| Death occurred in sleep environment, n (%) | ||||||
| Yes | 214 (89) | 878 (92) | 789 (90) | 127 (91) | 2008 (91) | .488 |
| No | 21 (9) | 63 (7) | 71 (8) | 7 (5) | 162 (7) | |
| Unknown/missing | 6 (2) | 13 (1) | 18 (2) | 5 (4) | 42 (2) | |
| Supervisor asleep at time of infant death, n (%) | ||||||
| Yes | 119 (49) | 432 (45) | 398 (45) | 84 (60) | 1033 (47) | <.001 |
| No | 54 (22) | 112 (12) | 140 (16) | 20 (14) | 326 (15) | |
| Unknown/missing | 68 (28) | 410 (43) | 340 (39) | 35 (25) | 853 (39) | |
| Supervisor impaired at time of infant death, n (%) | ||||||
| Yes | 49 (20) | 147 (15) | 178 (20) | 32 (23) | 406 (18) | <.001 |
| No | 148 (61) | 611 (64) | 591 (67) | 82 (59) | 1432 (65) | |
| Unknown/missing | 44 (18) | 196 (21) | 109 (12) | 25 (18) | 374 (17) | |
| Health care access | ||||||
| Insuranceb | ||||||
| Public/none | 167 (69) | 767 (80) | 609 (69) | 92 (66) | 1635 (74) | <.001 |
| Non-public | 22 (9) | 62 (6) | 164 (19) | 20 (14) | 268 (12) | |
| Unknown/missing | 52 (22) | 125 (13) | 105 (12) | 27 (19) | 309 (14) | |
| Barriers to prenatal care | ||||||
| Yes | 38 (16) | 137 (14) | 74 (8) | 21 (15) | 270 (12) | <.001 |
| No | 150 (62) | 548 (57) | 594 (68) | 92 (66) | 1384 (63) | |
| Unknown/missing | 53 (22) | 269 (28) | 210 (24) | 26 (19) | 558 (25) | |
| Child welfare involvement | ||||||
| Infant with open child protective services case | ||||||
| Yes | 28 (12) | 88 (9) | 97 (1) | 11 (8) | 224 (10) | .364 |
| No | 206 (85) | 853 (89) | 771 (88) | 124 (89) | 1954 (88) | |
| Unknown/missing | 7 (3) | 13 (1) | 10 (1) | 4 (3) | 34 (2) | |
| Infant maltreatment victim | ||||||
| Yes | 25 (10) | 131 (14) | 122 (14) | 21 (15) | 299 (14) | .753 |
| No | 200 (83) | 783 (82) | 719 (82) | 112 (81) | 1814 (82) | |
| Unknown/missing | 16 (7) | 40 (4) | 37 (4) | 6 (4) | 99 (5) | |
| Infant ever in out-of-home placement | ||||||
| Yes | * | * | * | * | 82 (4) | .796 |
| No | 230 (95) | 892 (94) | 815 (93) | 131 (94) | 2068 (93) | |
| Unknown/missing | * | * | * | * | 62 (3) | |
| Sibling ever in out-of-home placement | ||||||
| Yes | 20 (8) | 63 (7) | 91 (10) | 11 (8) | 185 (8) | .124 |
| No | 173 (72) | 683 (72) | 604 (69) | 102 (73) | 1562 (71) | |
| Not applicable | 22 (9) | 87 (9) | 98 (11) | 17 (12) | 224 (10) | |
| Unknown/missing | 26 (11) | 121 (13) | 85 (10) | 9 (6) | 241 (11) | |
| Caregiver with personal history of maltreatment victimization | ||||||
| At least 1 yes | 21 (9) | 142 (15) | 103 (12) | 15 (11) | 281 (13) | .005 |
| Both no | 56 (23) | 161 (17) | 201 (23) | 39 (28) | 457 (21) | |
| Both unknown or 1 no and other unknown | 164 (68) | 651 (68) | 574 (65) | 85 (61) | 1474 (67) | |
| Caregiver perpetrator of maltreatment | ||||||
| At least 1 yes | 52 (22) | 252 (26) | 210 (24) | 39 (28) | 553 (25) | .002 |
| Both no | 96 (40) | 265 (28) | 295 (34) | 57 (41) | 713 (32) | |
| Both unknown or 1 no and other unknown | 93 (39) | 437 (46) | 373 (42) | 43 (31) | 946 (43) | |
| Economic stability | ||||||
| Caregiver employmentc | ||||||
| At least 1 employed | 117 (49) | 385 (40) | 376 (43) | 75 (54) | 953 (43) | <.001 |
| Other | 29 (12) | 167 (18) | 83 (9) | 18 (13) | 297 (13) | |
| Both unknown | 95 (39) | 402 (42) | 419 (48) | 46 (33) | 962 (44) | |
| Substance use | ||||||
| Caregiver substance use | ||||||
| At least 1 yes | 66 (27) | 287 (30) | 346 (39) | 54 (39) | 753 (34) | <.001 |
| Both no | 61 (25) | 155 (16) | 176 (20) | 39 (28) | 431 (19) | |
| Both unknown or 1 no and other unknown | 114 (47) | 512 (54) | 356 (41) | 46 (33) | 1028 (46) | |
| Infant born drug exposed | ||||||
| Yes | 38 (16) | 105 (11) | 151 (17) | 17 (12) | 311 (14) | <.001 |
| No | 131 (54) | 408 (43) | 371 (42) | 77 (55) | 987 (45) | |
| Unknown/missing | 72 (30) | 441 (46) | 356 (41) | 45 (32) | 914 (41) | |
| Violence and criminal justice exposure | ||||||
| Caregiver victim or perpetrator of intimate partner violence | ||||||
| At least 1 yes | 35 (15) | 140 (15) | 93 (11) | 17 (12) | 285 (13) | <.001 |
| Both no | 67 (28) | 173 (18) | 248 (28) | 45 (32) | 533 (24) | |
| Both unknown or 1 no and other unknown | 139 (58) | 641 (67) | 537 (61) | 77 (55) | 1394 (63) | |
| Caregiver criminal history | ||||||
| At least 1 yes | 52 (22) | 286 (30) | 205 (23) | 42 (30) | 585 (26) | |
| Both no | 82 (34) | 188 (20) | 261 (30) | 44 (32) | 575 (26) | |
| Both unknown or 1 no and other unknown | 107 (44) | 480 (50) | 412 (47) | 53 (38) | 1052 (48) | <.001 |
| Housing and community | ||||||
| Infant ever homeless | ||||||
| Yes | * | * | * | * | 41 (2) | .804 |
| No | 207 (86) | 822 (86) | 739 (84) | 124 (89) | 1892 (86) | |
| Unknown/missing | * | * | * | * | * | |
| Incident area | ||||||
| Urban | 126 (52) | 586 (61) | 169 (19) | 56 (40) | 937 (42) | <.001 |
| Suburban | 75 (31) | 282 (30) | 274 (31) | 54 (39) | 685 (31) | |
| Rural/frontier | 31 (13) | 70 (7) | 391 (45) | 25 (18) | 517 (24) | |
| Unknown/missing | 9 (4) | 16 (2) | 44 (5) | 4 (3) | 73 (3) | |
Asterisk placed for any value <6 unless unknown/missing; P<.05 indicates statistical significance.
Centers for Disease Control and Prevention SUID and SDY Case Registry.
Dichotomized as public/no insurance vs non-public insurance. Self-pay was combined with non-public, and in cases where multiple insurances were public, insurance type was coded as public
When a caregiver was unemployed, on disability, stay-at-home, or retired, categorization was “other” versus employed or unknown.
Nearly one-third of NHB and one-fourth of Hispanic infants were premature; nearly one-third of NHB infants had low or very low birth weight (n=191, 20% and n=77, 8%, respectively). Half of childbearing parents of NHB infants experienced pregnancy complications (n=472, 49%). Statistically significant associations between infant race and ethnicity and prematurity, birth weight, childbearing parent smoking, prenatal care, and pregnancy complications were identified (P<.05) (Table 1).
Most deaths occurred in the sleep environment (n=2008, 91%). Nearly one-third of death categorizations in Sudden Death in the Young Registry were explained suffocation (n=633, 29%). The infant’s supervisor was impaired at time of death (n=406, 18%) for nearly 1 in 5 infants.
Distribution of Social Factors Across Infant Race and Ethnicity
Most infants who experienced SUID had public/no insurance (n=1635, 74%). Childbearing parents of NHB and Hispanic infants disproportionately experienced prenatal care barriers (n=137, 14% and n=38, 16%, respectively) (Table 1).
Family involvement with child welfare was common; 10% (n=224) of infants had open child protective services cases at time of death; 14% had histories of maltreatment victimization. For 13% of infants, at least 1 caregiver had personal history of maltreatment victimization; for 25%, at least 1 caregiver had history of maltreatment perpetration. Over one-fourth of NHB infants (n=252, 26%) had caregivers identified as maltreatment perpetrators. Nearly 10% of infants (n=185, 8%) had siblings who experienced out-of-home placement.
Overall, more than one-third of infants had a caregiver with substance use concerns (n=753, 34%). Thirty-nine percent (n=346) of NHW infants had a caregiver with substance use issues; nearly 1 in 5 (n=151, 17%) NHW infants were born prenatally exposed versus NHB (n=105, 11%), respectively.
Overall, few infants experienced homelessness (n=41, 2%). Nearly half of deaths occurred in urban locations (n=937, 42%). Nearly half of NHW infants died in rural areas (n=391, 45%); most NHB infants died in urban settings (n=586, 61%).
Statistically significant associations between infant race and ethnicity and insurance, caregiver criminal history, maltreatment perpetration and victimization, employment, substance use, and IPV were identified (Table 1).
Association of Social Factors with Cause of Death
Cause of death was undetermined (n=1292, 58%) or injury (n=636, 29%) for most infants; a minority of deaths were attributed to medical cause (n=284, 13%) (Table 2). Distribution of death cause across infant race and ethnicity is presented in Figure 1. Over one-third (n=226, 36%) of infants with injury-related deaths had caregivers with substance use concerns.
Table 2.
Distribution of social factors by cause of death, 2015–2020a
| Variables, N (%) | Cause of death | Adjusted odds ratio (of medical cause), adjusted for prematurity and birthweight | |||||||
|---|---|---|---|---|---|---|---|---|---|
|
|
|||||||||
| External cause of injury (n = 636) | Medical cause (n = 284) | Undetermined/ unknown (n = 1292) | Overall (n = 2212) | P value | Odds ratio | 95% CI | P value | ||
|
| |||||||||
| Infant race and ethnicity | |||||||||
| Non-Hispanic Black | 249 (30) | 117 (41) | 588 (46) | 954 (43) | .277 | Ref | |||
| Non-Hispanic White | 274 (43) | 109 (38) | 495 (38) | 878 (40) | 1.12 | 0.84 | 1.49 | .44 | |
| Hispanic | 67 (11) | 36 (13) | 138 (11) | 241 (11) | 1.32 | 0.86 | 1.97 | .19 | |
| Other non-Hispanic | 46 (7) | 22 (8) | 71 (6) | 139 (6) | 1.47 | 0.87 | 2.39 | .13 | |
| Health care access | |||||||||
| Insurance | |||||||||
| Public/none | 446 (70) | 194 (68) | 995 (77) | 1635 (74) | .007 | 1.691 | 1.178 | 2.389 | .004 |
| Non-public | 88 (14) | 47 (17) | 133 (10) | 268 (12) | Ref | ||||
| Unknown/missing | 102 (16) | 43 (15) | 164 (13) | 309 (14) | 1.137 | 0.761 | 1.666 | .52 | |
| Barriers to prenatal care | |||||||||
| Yes | 68 (11) | 25 (9) | 177 (14) | 270 (12) | .268 | Ref | |||
| No | 405 (64) | 180 (63) | 799 (62) | 1384 (63) | 1.555 | 1.014 | 2.482 | .052 | |
| Unknown/missing | 163 (26) | 79 (28) | 316 (24) | 558 (25) | 1.652 | 1.034 | 2.721 | .041 | |
| Child welfare involvement | |||||||||
| Infant with open child protective services case | |||||||||
| Yes | 56 (9) | 37 (13) | 131 (10) | 224 (10) | .52 | 0.741 | 0.511 | 1.099 | .124 |
| No | 569 (89) | 245 (86) | 1140 (88) | 1954 (88) | Ref | ||||
| Unknown/missing | 11 (2) | 2 (1) | 21 (2) | 34 (2) | 0.31 | 0.049 | 1.098 | .121 | |
| Infant maltreatment victim | |||||||||
| Yes | 79 (12) | 48 (17) | 172 (13) | 299 (14) | .532 | 0.787 | 0.562 | 1.119 | .170 |
| No | 532 (84) | 226 (80) | 1056 (82) | 1814 (82) | Ref | ||||
| Unknown/missing | 25 (4) | 10 (4) | 64 (5) | 99 (4) | 0.628 | 0.287 | 1.259 | .213 | |
| Infant ever in out-of-home placement | |||||||||
| Yes | 19 (3) | 19 (7) | 44 (3) | 82 (4) | .119 | Ref | |||
| No | 601 (94) | 260 (92) | 1207 (93) | 2068 (93) | 0.5 | 0.298 | 0.877 | .011 | |
| Unknown/missing | 16 (3) | 5 (2) | 41 (3) | 62 (3) | 0.298 | 0.093 | 0.804 | .025 | |
| Sibling ever in out-of-home placement | |||||||||
| Yes | 54 (8) | 17 (6) | 114 (9) | 185 (8) | .681 | Ref | |||
| No | 456 (72) | 197 (69) | 909 (70) | 1562 (71) | 1.422 | 0.865 | 2.486 | .188 | |
| Unknown/missing | 70 (11) | 32 (11) | 139 (11) | 241 (11) | 1.458 | 0.787 | 2.788 | .240 | |
| Not applicable | 56 (9) | 38 (13) | 130 (10) | 224 (10) | 2.04 | 1.12 | 3.852 | .023 | |
| Caregiver victim of child maltreatment | |||||||||
| At least 1 yes | 87 (14) | 36 (13) | 158 (12) | 281 (13) | .257 | 0.965 | 0.649 | 1.399 | .854 |
| Both no | 149 (23) | 47 (17) | 261 (20) | 457 (21) | 0.737 | 0.52 | 1.026 | .078 | |
| Both unknown or 1 no and other unknown | 400 (63) | 201 (71) | 873 (68) | 1474 (67) | Ref | ||||
| Caregiver perpetrator of maltreatment | |||||||||
| At least 1 yes | 165 (26) | 66 (23) | 322 (25) | 553 (25) | .1 | 0.852 | 0.616 | 1.169 | .327 |
| Both no | 231 (36) | 87 (31) | 395 (31) | 713 (32) | 0.898 | 0.668 | 1.202 | .472 | |
| Both unknown or 1 no and other unknown | 240 (38) | 131 (46) | 575 (45) | 946 (43) | Ref | ||||
| Economic stability | |||||||||
| Caregiver employment | |||||||||
| At least 1 employed | 284 (45) | 108 (38) | 561 (43) | 953 (43) | .562 | 0.792 | 0.602 | 1.04 | .094 |
| Other | 76 (12) | 41 (14) | 180 (14) | 297 (13) | 0.968 | 0.654 | 1.406 | .866 | |
| Both unknown | 276 (43) | 135 (48) | 551 (43) | 962 (44) | Ref | ||||
| Substance use | |||||||||
| Caregiver substance use | |||||||||
| At least 1 yes | 226 (36) | 70 (25) | 457 (35) | 753 (34) | .031 | 0.627 | 0.461 | 0.847 | .003 |
| Both no | 126 (20) | 67 (24) | 238 (18) | 431 (19) | 1.143 | 0.829 | 1.563 | .408 | |
| Both unknown or 1 no and other unknown | 284 (45) | 147 (52) | 597 (46) | 1028 (46) | Ref | ||||
| Infant born drug exposed | |||||||||
| Yes | 91 (14) | 35 (12) | 185 (14) | 311 (14) | .987 | Ref | |||
| No | 281 (44) | 132 (46) | 574 (44) | 987 (45) | 1.212 | 0.821 | 1.833 | .346 | |
| Unknown/missing | 264 (42) | 117 (41) | 533 (41) | 914 (41) | 1.159 | 0.78 | 1.76 | .476 | |
| Violence and criminal justice exposure | |||||||||
| Caregiver victim or perpetrator of intimate partner violence | |||||||||
| At least 1 yes | 68 (11) | 27 (10) | 190 (15) | 285 (13) | .029 | 0.658 | 0.421 | 0.991 | .054 |
| Both no | 175 (28) | 63 (22) | 295 (23) | 533 (24) | 0.857 | 0.627 | 1.158 | .323 | |
| Both unknown or 1 no and other unknown | 393 (62) | 194 (68) | 807 (62) | 1394 (63) | Ref | ||||
| Caregiver criminal history | |||||||||
| At least 1 yes | 161 (25) | 62 (22) | 362 (28) | 585 (26) | .429 | 0.732 | 0.529 | 1.001 | .054 |
| Both no | 174 (27) | 74 (26) | 327 (25) | 575 (26) | 0.913 | 0.672 | 1.232 | .557 | |
| Both unknown or 1 no and other unknown | 301 (47) | 148 (52) | 603 (47) | 1052 (48) | Ref | ||||
| Housing and community | |||||||||
| Infant ever homeless | |||||||||
| Yes | * | * | * | 41 (2) | .691 | 1.369 | 0.539 | 4.627 | .556 |
| No | 536 (84) | 245 (86) | 1111 (86) | 1892 (86) | Ref | ||||
| Unknown/missing | * | * | * | 279 (13) | 1.353 | 0.499 | 4.743 | .59 | |
| Incident area | |||||||||
| Urban | 243 (38) | 117 (41) | 577 (45) | 937 (42) | .398 | Ref | |||
| Suburban | 203 (32) | 81 (29) | 401 (31) | 685 (31) | 1.125 | 0.841 | 1.511 | .431 | |
| Rural | 165 (26) | 77 (27) | 271 (21) | 517 (24) | 0.839 | 0.618 | 1.143 | .262 | |
| Frontier | * | * | * | * | * | * | * | * | |
| Unknown/missing | * | * | * | * | 0.821 | 0.443 | 1.644 | .551 | |
Asterisk placed for any value <6 unless unknown/missing; P<.05 indicates statistical significance.
Centers for Disease Control and Prevention SUID and SDY Case Registry.
Figure 1.
Distribution of cause of death across infant race and ethnicity.
In adjusted logistic regression analysis, caregiver criminal history, IPV, and substance use were associated with lower odds of medical cause of death versus injury and undetermined causes (Table 2). Infant race and ethnicity were not associated with cause of death.
Social Factors Across Death Groupings
Few deaths (n=311) were categorized as Group 2. No statistically significant differences in social factors were identified across groupings (Table 3).
Table 3.
Distribution of factors across groups, 2015–2020a
| Variables, n (%) | Group 2 | Group 1 | Overall | P value | Adjusted OR (for medical cause), adjusted for prematurity and birthweight | 95% CI | P value | |
|---|---|---|---|---|---|---|---|---|
|
| ||||||||
| (n = 311) | (n = 1901) | (n = 2212) | ||||||
|
| ||||||||
| Infant race and ethnicity | ||||||||
| Non-Hispanic Black | 136 (44) | 818 (43) | 954 (43) | .965 | Ref | |||
| Non-Hispanic White | 116 (37) | 762 (40) | 878 (40) | 0.989 | 0.752 | 1.3 | .934 | |
| Hispanic | 36 (12) | 205 (11) | 241 (11) | 0.902 | 0.608 | 1.367 | .617 | |
| Other non-Hispanic | 23 (7.4) | 116 (6.1) | 139 (6.3) | 0.776 | 0.483 | 1.292 | .31 | |
| Health care access | ||||||||
| Insurance | ||||||||
| Public/none | 221 (71) | 1414 (74) | 1635 (74) | 1.273 | 0.877 | 1.813 | .191 | |
| Non-public | 42 (14) | 226 (12) | 268 (12) | .821 | Ref | |||
| Unknown/missing | 48 (15) | 261 (14) | 309 (14) | 1.131 | 0.699 | 1.829 | .616 | |
| Barriers to prenatal care | ||||||||
| Yes | 33 (11) | 237 (13) | 270 (12) | .876 | Ref | |||
| No | 194 (62) | 1190 (63) | 1384 (63) | 0.794 | 0.503 | 1.302 | .341 | |
| Unknown/missing | 84 (27) | 474 (25) | 558 (25) | 0.932 | 0.535 | 1.715 | .813 | |
| Child welfare involvement | ||||||||
| Infant with open child protective services case | ||||||||
| Yes | 36 (12) | 188 (10) | 224 (10) | 0.879 | 0.606 | 1.307 | .51 | |
| No | 273 (88) | 1681 (88) | 1954 (88) | .619 | Ref | |||
| Unknown/missing | 2 (1) | 32 (2) | 34 (2) | 2.707 | 0.807 | 16.846 | .176 | |
| Infant maltreatment victim | ||||||||
| Yes | 52 (17) | 247 (13) | 299 (14) | 0.799 | 0.578 | 1.123 | .186 | |
| No | 248 (80) | 1566 (82) | 1814 (82) | .45 | Ref | |||
| Unknown/missing | 11 (4) | 88 (5) | 99 (5) | 1.244 | 0.679 | 2.513 | .509 | |
| Infant ever in out-of-home placement | ||||||||
| Yes | 19 (6) | 63 (3) | 82 (4) | .0709 | Ref | |||
| No | 288 (93) | 1780 (94) | 2068 (94) | 1.759 | 1.005 | 2.946 | .039 | |
| Unknown/missing | 4 (1) | 58 (3) | 62 (3) | 4.215 | 1.467 | 15.285 | .014 | |
| Sibling ever in out-of-home placement | ||||||||
| Yes | 24 (8) | 161 (9) | 185 (8) | .987 | Ref | |||
| No | 217 (70) | 1345 (71) | 1562 (71) | 0.921 | 0.571 | 1.428 | .725 | |
| Unknown/missing | 34 (11) | 207 (11) | 241 (11) | 0.935 | 0.525 | 1.644 | .817 | |
| Not Applicable | 36 (12) | 188 (10) | 224 (10) | 0.755 | 0.432 | 1.337 | .352 | |
| Caregiver victim of child maltreatment | ||||||||
| At least 1 yes | 41 (13) | 240 (13) | 281 (13) | 0.945 | 0.662 | 1.376 | .76 | |
| Both no | 59 (19) | 398 (21) | 457 (21) | 1.103 | 0.812 | 1.518 | .538 | |
| Both unknown or 1 no and other unknown | 211 (68) | 1263 (66) | 1474 (67) | .958 | Ref | |||
| Caregiver perpetrator of maltreatment | ||||||||
| At least 1 yes | 85 (27) | 468 (25) | 553 (25) | 0.837 | 0.621 | 1.133 | .247 | |
| Both no | 100 (32) | 613 (32) | 713 (32) | 0.903 | 0.679 | 1.203 | .483 | |
| Both unknown or 1 no and other unknown | 126 (41) | 820 (43) | 946 (43) | .875 | Ref | |||
| Economic stability | ||||||||
| Caregiver employment | ||||||||
| At least 1 employed | 131 (42) | 822 (43) | 953 (43) | 1.022 | 0.787 | 1.327 | .871 | |
| Other (stay at home or retired) | 43 (14) | 254 (13) | 297 (13) | 0.989 | 0.685 | 1.454 | .956 | |
| Both unknown/missing | 137 (44) | 825 (43) | 962 (44) | .997 | Ref | |||
| Substance use | ||||||||
| Caregiver substance use | ||||||||
| At least 1 yes | 87 (28) | 666 (35) | 753 (34) | 1.3 | 0.98 | 1.733 | .071 | |
| Both no | 72 (23) | 359 (19) | 431 (20) | 0.834 | 0.614 | 1.14 | .248 | |
| Both unknown or 1 no and other unknown | 152 (49) | 876 (46) | 1028 (47) | .144 | Ref | |||
| Infant born drug exposed | ||||||||
| Yes | 37 (12) | 274 (14) | 311 (14) | .491 | Ref | |||
| No | 153 (49) | 834 (44) | 987 (45) | 0.726 | 0.487 | 1.059 | .106 | |
| Unknown/missing | 121 (39) | 793 (42) | 914 (41) | 0.872 | 0.58 | 1.284 | .498 | |
| Violence and criminal justice exposure | ||||||||
| Caregiver victim or perpetrator of intimate partner violence | ||||||||
| At least 1 yes | 30 (10) | 255 (13) | 285 (13) | 1.335 | 0.898 | 2.047 | .168 | |
| Both no | 88 (28) | 445 (23) | 533 (24) | 0.781 | 0.593 | 1.034 | .081 | |
| Both unknown or 1 no and other unknown | 193 (62) | 1201 (63) | 1394 (63) | .227 | Ref | |||
| Caregiver criminal history | ||||||||
| At least 1 yes | 76 (24) | 509 (27) | 585 (26) | 1.046 | 0.776 | 1.418 | .772 | |
| Both no | 91 (29) | 484 (26) | 575 (26) | 0.828 | 0.622 | 1.106 | .197 | |
| Both unknown or 1 no and other unknown | 144 (46) | 908 (48) | 1052 (48) | .707 | Ref | |||
| Housing and community | ||||||||
| Infant ever homeless | ||||||||
| Yes | 7 (2) | 34 (2) | 41 (2) | 0.779 | 0.36 | 1.945 | .557 | |
| No | 265 (86) | 1627 (86) | 1892 (86) | .989 | Ref | |||
| Unknown/missing | 39 (13) | 240 (13) | 279 (13) | 0.986 | 0.692 | 1.438 | .939 | |
| Incident area | ||||||||
| Urban | 133 (43) | 804 (42) | 937 (42) | .965 | Ref | |||
| Suburban | 87 (28) | 598 (32) | 685 (31) | 1.125 | 0.841 | 1.511 | .431 | |
| Rural | * | * | * | 0.839 | 0.618 | 1.143 | .262 | |
| Frontier | * | * | * | |||||
| Unknown/missing | 12 (4) | 61 (3) | 73 (3) | 0.821 | 0.443 | 1.644 | .551 | |
Asterisk placed for any value <6 unless unknown/missing; P<.05 indicates statistical significance.
Centers for Disease Control and Prevention SUID and SDY Case Registry.
Discussion
Understanding root causes and conditions contributory to several thousand SUID annually in the U.S. persists as a public health priority. Prevention emphasis has historically appropriately focused on the infant sleep environment, primarily through reduction of individual level hazards. While contributing to overall decline in SUID rate, emerging racial disparities among historically marginalized populations facing adverse social factors suggest need for paradigm shift toward broader consideration of social factor impact on SUID risk and social policy change as SUID prevention strategy. Harnessing multi-jurisdictional data from Centers for Disease Control and Prevention’s SUID and Sudden Death in the Young Case Registry with racial, ethnic, and geographic diversity, this study characterizes social factors across infants who experience sudden death, examining distribution of social factors across infant race and ethnicity, association of social factors with causes of death and investigator-derived death groupings that may meaningfully inform future research, advocacy, and prevention efforts.
Adverse social factors appear prevalent among infants who die suddenly and unexpectedly. Across infants included in the multi-jurisdictional data set, most had public or no insurance, a proxy for socioeconomic disadvantage. Substance use among caregivers of infants who experienced SUID was frequent; more than one-third of infants had at least 1 caregiver with CDR-determined substance use concerns. Substance use may reduce/disrupt parenting capacity through impairment, impact decision-making, supervision, and contribute to risk-taking behavior including related to sleep environment hazards. Substance use has strong inter-relationship with other social factors including violence exposure and criminal justice system involvement that may impact socioeconomic disadvantage and potentiate risk-taking behaviors including involving infant care.22,23 Over one-third of infants whose deaths were attributed to injury in this study (n=226, 36% including sleep-related suffocation, drowning) had caregivers with CDR-determined substance use concerns. Several studies have previously shown caregiver alcohol and/or illicit substance use in combination with surface sharing while impaired place infants at particularly high risk for unintentional suffocation, considered a type of injury-related death.8,24,25 Caregivers who use substances may thus represent a vulnerable population in need of targeted, secondary SUID prevention efforts by health care professionals to reduce injury-related deaths.
When examining association of social factors with SUID cause, caregiver substance use, criminal justice system involvement, and experience of IPV were associated with lower odds of medical cause of death versus other causes. While meriting further study, these social factors may reflect risk-taking behaviors that specifically heighten injury-related death potential versus death from medical conditions. Importantly, these social factors may also compound socioeconomic disadvantage in varied ways that limit access to employment, material resources, education, and health care, preventing opportunities for safe sleep reinforcement thereby potentiating SUID risk. Like those affected by family-level substance use adversity, infants from families affected by criminal justice system involvement and violence may similarly represent vulnerable populations that could benefit from targeted, secondary SUID prevention efforts, including potentially material supports, referrals to home visitors, and other community-based programming.
Recent national data suggest infants from NHB families disproportionately experience SUID.2,3 In this study, notably no statistical associations were identified across infant race and ethnicity, cause of death, or investigator-derived death grouping among analyzed SUID. This raises concern social factors influence and may even underlie root causes of emerging SUID disparities in national fatality data. Social factors may be clustered or concentrated in marginalized populations through decades of embedded structural racism that perpetuate inequities in health outcomes and contribute to observed widening disparities in SUID rate. Prematurity and low birth weight, disproportionately prevalent among NHB infants, are leading causes of infant mortality nationally and risk factors associated with SUID; both prematurity and low birth weight are hypothesized to have relationship with social factors like socioeconomic disadvantage, barriers to adequate housing, education, and health care, frequently compounded by structural racism.7,10–12,26–29 In this study, whereas one-fourth of infants overall were premature (n=553, 25%), NHB infants experienced disproportionate prematurity and low/very low birth weights. Nearly half of childbearing parents of NHB infants experienced pregnancy complications (n=472, 49%). Statistically significant associations between infant race and ethnicity and histories of prematurity, birth weight, prenatal care, and pregnancy complications among infants who later experienced SUID were identified. Results suggest clustering of multiple SUID risk factors among infants from marginalized NHB families, supporting concerns social conditions–including limited access to adequate nutrition, health care, stable housing, and education–may be potentiating risk among this population and contribute to, or event underlie, disparities in SUID rate.
Most sudden unexpected deaths in this study were associated with the sleep environment, a persistent trend over several decades despite robust prevention efforts. Results of this study raise concern public messaging campaigns like “Safe to Sleep” (formerly “Back to Sleep”) and individual-level anticipatory guidance delivered in health care settings may fail to adequately, equitably reach vulnerable populations in need of education, whose health care access is limited by poverty, transportation barriers, lack of insurance or other social factors. Even when messaging is delivered, co-morbid social factors like substance use and poverty may prevent infant placement in safe sleep environments (either related to substance impairment or paucity of material resources, limited access to cribs/safe sleep spaces). Safe sleep recommendations at the individual level may thus lack prevention power if unable to be heeded due to social factors experienced by the infant’s family, failing to remedy underlying root causes of SUID risk. Further study of these relationships, specifically intersect of infant sleep circumstances with social factors, is needed to inform novel prevention efforts.4
Children’s welfare has long been tied to social factors that can only meaningfully improve through society-wide, social policy reforms.30 Contribution of social factors to SUID risk, first described as the Triple Risk Hypothesis in 1994, is thus not a novel concept.17,31,32 A paradigm shift toward enhanced emphasis on the relationship between social factors and SUID risk is urgently needed within prevention frameworks (Figure 2). Broader community and society-level interventions including policy change that address social conditions and inequities experienced by vulnerable families to meaningfully impact infant health outcomes is necessary.2,3,33 Addressing social factors through policy change may reduce SUID and should augment current SUID prevention efforts. Existing social policies in the U.S. targeting poverty, childbearing parents, and marginalized populations have long been critiqued as woefully inadequate; despite substantial income inequality, the U.S. lacks paid maternity leave, universal health care, and limits federal assistance programs for impoverished families.22 Social policy change that allocates additional resources to facilities treating substance use among caregivers with children may also reduce SUID risk, given noted associations of substance use with risk-taking behaviors including bedsharing and placement in unsafe sleep environments that heighten injury-related death risk.8,9,25 Broad-sweeping social policy change that addresses supplemental income programs, minimum wage policies, nutrition safeguard programs, affordable health care coverage, immigration-related policies, reproductive and family-based policies (including childcare access) are necessary to improve population health inequities34 and dismantle impact of structural racism on SUID risk. Reduction of funding and elimination of public health programs focused on safe sleep initiatives threaten to further exacerbate existing disparities in SUID rate and reduce opportunities for counseling and education of families about sleep environment risk.
Figure 2.
Social factors and SUID - an adaptation of the Triple Risk Hypothesis.11
Several study limitations exist. Importantly, cross-sectional design limits conclusions regarding causation of SUID and social factors. While Case Registry data is multi-jurisdictional, geographically diverse, and compiled across a variety of clinical and investigative sources during the CDR process, data availability, accessibility, and consistency contributes to multiple social factor variables lacking large proportions of data. As imputation could not be performed, to mitigate impact of missing data and reduce data quality concerns, social factor data involving negative responses by single caregivers were grouped with unknown responses, whereas a single caregiver affirmative response was considered a positive exposure for the social factor. This approach may have underestimated the true proportion of affirmative responses for each social factor across caregivers. Additionally, missing data may account for lack of statistical association between SUID and cause of death for multiple social factor variables. Assessment of additional social factors, such as the interplay of food insecurity, household crowding, and unsafe sleep and inclusion of both risk and protective factors, may valuably inform future prevention strategies. Thirteen U.S. jurisdictions contributed data to this study, which may impact overall representativeness. Related to the NFR-CRS data tool itself, within the tool SIDS is categorized as a medical cause of death, like cancer and infection; jurisdictional variability in death classification and recent diagnostic shift away from SIDS designation and toward suffocation and undetermined causes may have impacted cause of death categorization during the CDR process. Nearly one-third of cases were attributed to suffocation, raising concern that presence of unsafe sleep factors may frequently be considered causal; few differences are noted across accidental and unexplained deaths, suggesting similarity. Further study of the impact of investigative practices on cause of death categorization is needed.
Despite these appreciable limitations, this novel effort represents 1 of only several studies at present specifically aiming to further understanding of relationships across social factors and SUID and responds to growing concern that social factors may underlie emerging disparities in SUID data. Utilizing a multi-jurisdictional diverse dataset, this study importantly identifies gaps in existing data infrastructure that must be addressed to meaningfully assess contribution of social factors to SUID risk, including efforts to improve data quality and reduce large proportions of missing social factor data. Future directions for research, advocacy, and SUID prevention efforts include development and implementation of improved data collection efforts regarding social factors and funding support among participating jurisdictions to improve data quality across research and policy analyses.
Conclusion
Utilizing a multi-jurisdictional SUID Case Registry, relationships between social factors and cause of death among infants experiencing SUID were identified; social factors including caregiver criminal history, substance use, and intimate partner violence were associated with lower odds of medical cause of death versus other causes. Distribution of social factors among infants experiencing SUID, which may contribute to emerging disparities, requires further study, and efforts are needed to improve data quality around social factors. SUID prevention efforts should address impact of social factors on fatality vulnerability and advocacy for social policy change.
What’s New.
Emerging disparities compel a paradigm shift from narrow focus on the safe sleep environment to broader social inequities. Social context and family vulnerability factors impacting SUID across racial/ethnic groups and cause of death are compared to identify targeted prevention opportunities.
Acknowledgments
We thank Sudden Death in the Young and SUID Case Registry participants, the leadership and data team at the National Center for their support in preparing data from the NFR-CRS and review of this manuscript, Drs. Matthew Di Guglielmo and Claudine Jurkovitz for their feedback, and Kim Eissmann and Michelle Stofa for their efforts editing the manuscript.
Funding Source
Work supported in part by an institutional development Award (IDeA) from the National Institutes of Health under grant number U54-GM104941 (PI: Hicks). The dataset was provided by the National Center for Fatality Review and Prevention (National Center), which is funded in part by Cooperative Agreement Number UG7MC28482 from the U.S. Department of Health and Human Services (HHS), Health Resources and Services Administration (HRSA), Maternal and Child Health Bureau (MCHB) as part of an award totaling $5,149,996 annually with 0 percent financed with non-governmental sources. The contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Center, HHS, or the participating states.
Role of Funder/Sponsor:
The NIH had no role in the design and conduct of the study.
Footnotes
Conflict of Interest Disclosure: The authors have no conflicts of interest relevant to this article to disclose.
Credit Author Statement
Conceptualization: SAD; Data curation: SAD, CED and JH; Formal analysis: SAD, CEL and JH; Investigation: SAD; Methodology: SAD, CEL, JH and ADJ; Project administration: ADJ; Resources: CEL and JH; Software: CEL and JH; Supervision: SAD and ADJ; Validation: SAD, CEL, JH and ADJ; Visualization: SAD; Writing—original draft: SAD; Writing—review & editing: SAD, CEL, JH and ADJ.
Declaration of interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Clinical Trial Registration
N/A
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Centers for Disease Control and Prevention. Data and Statistics for SUID and SIDS. 2024. (https://www.cdc.gov/sudden-infant-death/data-research/data) Accessed November 5, 2024.
- 2.Moon RY, Carlin RF, Hand I Evidence base for 2022 updated recommendations for a safe infant sleeping environment to reduce the risk of sleep-related infant deaths. Pediatrics. 2022; 150:e2022057991. [DOI] [PubMed] [Google Scholar]
- 3.Shapiro-Mendoza CK, Woodworth KR, Cottengim CR, et al. Sudden unexpected infant deaths: 2015–2020. Pediatrics. 2023; 151:e2022058820. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Boyer BT, Lowell GS, Roehler DR, et al. Racial and ethnic disparities of sudden unexpected infant death in large US cities: a descriptive epidemiological study. Inj Epidemiol. 2022; 9:12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Carlin RF, Hauck FR, Moon RY Increasing disparities in sudden unexpected infant deaths reflect societal failures. Pediatrics. 2023; 151:e2022060798. [DOI] [PubMed] [Google Scholar]
- 6.Hwang SS, Melvin P, Diop H, et al. Implementation of safe sleep practices in Massachusetts NICUs: a state-wide QI collaborative. J Perinatol. 2018; 38:593–9 [DOI] [PubMed] [Google Scholar]
- 7.Lorch SA, Enlow E The role of social determinants in explaining racial/ethnic disparities in perinatal outcomes. Pediatr Res. 2016; 79:141–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Blair PS, Sidebotham P, Evason-Coombe C, et al. Hazardous cosleeping environments and risk factors amenable to change: case-control study of SIDS in south west England. BMJ. 2009; 339:b3666. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Garstang JJ, Sidebotham P Qualitative analysis of serious case reviews into unexpected infant deaths. Arch Dis Child. 2019; 104:30–6 [DOI] [PubMed] [Google Scholar]
- 10.Hwang SS, Bourque SL, Hannan KE, et al. Racial and ethnic disparities in sudden unexpected infant death among US infants born preterm. J Pediatr. 2023; 260:113498. [DOI] [PubMed] [Google Scholar]
- 11.Matoba N, Suprenant S, Rankin K, et al. Mortgage discrimination and preterm birth among African American women: an exploratory study. Health Place. 2019; 59;102193. [DOI] [PubMed] [Google Scholar]
- 12.Salow AD, Pool LR, Grobman WA, et al. Associations of neighborhood level racial residential segregation with adverse pregnancy outcomes. Am J Obstet Gynecol. 2018; 218:351.e1–.e7 [DOI] [PubMed] [Google Scholar]
- 13.Burns KM, Bienemann L, Camperlengo L, et al. The Sudden Death in the Young Case Registry: collaborating to understand and reduce mortality. Pediatrics. 2018; 139:e20162757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Covington TM The US national child death review case reporting system. Inj Prev. 2011; 17:i34–7 [DOI] [PubMed] [Google Scholar]
- 15.National Center for Fatality Review and Prevention. National center program manual. 2021. (https://ncfrp.org/wp-content/uploads/NCRPCD-Docs/ProgramManual.pdf) Accessed February 28, 2024.
- 16.Data Coordinating Center. SUID and SDY Case Registry. 2025. (https://sdyregistry.org/tools-for-providers/) Accessed February 28, 2024.
- 17.Moon RY, Fu L Sudden infant death syndrome: an update. Pediatr Rev. 2012; 33:314–20 [DOI] [PubMed] [Google Scholar]
- 18.National CRFP. The National Fatality Review-Case Reporting System. 2025. (https://ncfrp.org/data/nfr-crs/) Accessed November 15, 2023.
- 19.SDY-Categorization-Algorithm_V10. SUID and SDY Case Registry. 2024. (https://sdyregistry.org/wp-content/uploads/SDY-Categorization-Algorithm_V10_Clean.pdf) Accessed November 15, 2023. [Google Scholar]
- 20.National Center for Fatality Review and Prevention. Data dictionary version 6.0. National Fatality Review Case Reporting System. 2024. (https://ncfrp.org/wp-content/uploads/DataDictionary_CRS_v6-0.pdf) Accessed November 15, 2023. [Google Scholar]
- 21.R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. (https://www.R-project.org/) Accessed November 15 2023. [Google Scholar]
- 22.Bartick M, Tomori C Sudden infant death and social justice: a syndemics approach. Matern Child Nutr. 2019; 15:e12652. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Khoury L, Tang YL, Bradley B, et al. Substance use, childhood traumatic experience, and posttraumatic stress disorder in an urban civilian population. Depress Anxiety. 2010; 27:1077–86 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Carpenter RG, Irgens LM, Blair PS, et al. Sudden unexplained infant death in 20 regions in Europe: case control study. Lancet. 2004; 363:185–91 [DOI] [PubMed] [Google Scholar]
- 25.Erck Lambert AB, Shapiro-Mendoza CK, Parks SE, et al. Characteristics of sudden unexpected infant deaths on shared and nonshared sleep surfaces. Pediatrics. 2024; 153:e2023061984. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Bishop-Royce J, Lange-Maia B, Murray L, Shah RC, et al. Structural racism, socioeconomic marginalization, and infant mortality. Public Health. 2021; 190:55–61 [DOI] [PubMed] [Google Scholar]
- 27.Centers for Disease Control and Prevention. Injury prevention & control. Injuries among children and teens. 2024. (https://www.cdc.gov/child-passenger-safety/about/?CDC_AAref_Val=https://www.cdc.gov/injury/features/child-injury/index.html) Accessed November 15 2023.
- 28.Centers for Disease Control and Prevention. Reproductive health. Infant mortality. 2024. (https://www.cdc.gov/maternal-infant-health/infant-mortality/?CDC_AAref_Val=https://www.cdc.gov/reproductivehealth/maternalinfanthealth/infantmortality.htm) Accessed November 15, 2023.
- 29.Jang CJ, Lee HC A review of racial disparities in infant mortality in the US. Children (Basel). 2022; 9:257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Roberts DE Shattered Bonds: the Color of Child Welfare. Penn Carey Law University of Pennsylvania, Philadelphia, PA, 2001 [Google Scholar]
- 31.Filiano JJ, Kinney HC A perspective on neuropathologic findings in victims of the sudden infant death syndrome: the triple risk model. Biol Neonate. 1994; 65:194–7 [DOI] [PubMed] [Google Scholar]
- 32.Perrone S, Lembo C, Moretti S, et al. Sudden infant death syndrome: beyond risk factors. Life (Basel). 2021; 11:184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Moon RY, Hauck FR, Colson ER, et al. The effect of nursing quality improvement and mobile health interventions on infant sleep practices: a randomized clinical trial. JAMA. 2017; 318:351–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Clark EC, Cranston E, Polin T, et al. Structural interventions that affect racial inequities and their impact on population health outcomes: a systematic review. BMC Public Health. 2022; 22:2162. [DOI] [PMC free article] [PubMed] [Google Scholar]